Surface Crack Detection and Localisation for Automated Underground Mine Void Characterisation

Why it matters

Safety is an important aspect in underground mining operations that mining companies put high priority in. In comparison to other safety hazards, geological failure (e.g. roof collapse, rockfalls, etc) contributes to the highest number of fatalities and injuries in underground mines. Furthermore, these undesirable events can also cause equipment breakdowns that lead to company downtime and a loss in revenue. To prevent such hazard, geotechnical engineers conduct regular surveys of the underground tunnels and voids to detect early signs of geological failures and hence design and implement support systems. The practice of underground surveying is a risky operation due to the unpredictable and harsh underground conditions. By deploying an unmanned vehicle equipped with vision sensors for automated early signs detection, the risk on human can be eliminated.

Even though there are a number of early signs for geological failure, as a first step, this project focuses on the automatic detection of surface cracks (a.k.a fracture or sharp deformation breaks).

Overview

This project is the PhD of Faris Azhari, supervised by A/Prof. Thierry Peynot (QUT) and Dr. Charlotte Sennersten (CSIRO), and partly supported by Mining3.

This project focuses on the detection of surface cracks which includes:

• A review of state-of-the-art methods and technology used in crack detection including practices in different environmental contexts.
• Data collection of rock surfaces in underground and above ground environment using different sensors.
• Testing and evaluating existing automated crack detection systems on data collected, including identification of their limitations.
• Development of novel method(s) to reliably detect cracks in underground mine walls using LIDAR and/or vision.